Skip to content
AI automation examples · Australia

AI automation examples with dummy data

See how a workflow runs, step by step, with a person checking what matters.

Demonstration with dummy data

Every example on this page uses made-up names and no figures. They show how we work. They are not client results and not a measured saving. Each one uses the same five marks: trigger, AI step, check by a person, action and record. Back to AI automation.

Example 1

An enquiry is sorted, a reply is prepared and the record is updated

A website enquiry arrives, the AI sorts it and drafts a reply, a person approves it, and the customer record is updated.

  1. 01 · Trigger

    An enquiry arrives

    A website form is sent.

  2. 02 · AI step

    The AI sorts and drafts

    It picks the kind of enquiry and writes a draft from approved wording.

  3. 03 · Check by a person

    A person approves

    Nothing is sent before a named person approves or edits the draft.

  4. 04 · Action

    The reply is sent

    The approved reply goes out and a follow-up reminder is set.

  5. 05 · Record

    The record is updated

    The enquiry, the reply and the reminder are saved on the customer record.

An enquiry is sorted, a reply is prepared and the record is updated. Demonstration with dummy data.

Demonstration with dummy data

Enquiry summary and draft reply

Enquiry type
Quote request.
From
Sam Example, Example Pty Ltd.
Draft reply
Hi Sam, thanks for getting in touch. We have your request and will send a quote after a short call. What time suits you this week?
Needs your decision
No phone number was given. Approve, edit or ask the customer for a number.

Made-up names and numbers. This is not a client result.

Used on: AI automation (main page) · Enquiry and lead follow-up automation · Workflow and business process automation

Example 2

A quote draft is prepared and a person edits and sends it

A quote draft is built from an enquiry and a made-up price list. The person edits it and sends it.

  1. 01 · Trigger

    A quote request arrives

    An enquiry asks for a quote for Item A and Item B.

  2. 02 · AI step

    A draft is prepared

    The items are matched to a made-up list and the draft text is written.

  3. 03 · Check by a person

    A person edits and approves

    The person checks every line and changes anything that is wrong.

  4. 04 · Action

    The quote is sent

    The approved quote is sent from the person's own mailbox.

  5. 05 · Record

    The quote is recorded

    The version that was sent is saved on the customer record.

A quote draft is prepared and a person edits and sends it. Demonstration with dummy data.

Demonstration with dummy data

Quote draft (made-up example)

Customer
Sam Example, Example Pty Ltd.
Lines
Item A, quantity 1. Item B, quantity 2. Amounts are filled in from the real price list when it is used.
Flag
Item B has two options in the list. The person chooses one before sending.

Made-up names and numbers. This is not a client result.

Used on: Quote and proposal automation

Example 3

An invoice is read and checked, and a draft entry is prepared

A PDF invoice is read, its fields are checked against rules and a draft entry is prepared for approval.

  1. 01 · Trigger

    An invoice arrives

    A PDF reaches a shared mailbox.

  2. 02 · AI step

    The fields are read and checked

    Supplier, number, lines and total are read and compared with the order.

  3. 03 · Check by a person

    A person approves

    Anything unclear waits for a person. Nothing is paid by the system.

  4. 04 · Action

    A draft entry is prepared

    A draft goes into the accounts software for approval.

  5. 05 · Record

    The run is recorded

    The file, the checks and the decision are saved.

An invoice is read and checked, and a draft entry is prepared. Demonstration with dummy data.

Demonstration with dummy data

Invoice check (made-up example)

Supplier
Example Office Supplies. Found on the supplier list.
Lines read
Item A, Item B. Both match the order lines.
Flag
The due date is missing on the PDF. It is marked for the approver to confirm.

Made-up names and numbers. This is not a client result.

Used on: Invoice and accounts payable automation · Document processing and data entry automation · RPA services

Example 4

A support inbox: messages are sorted and answers are drafted from an approved list

Messages are sorted, answers are drafted from an approved answer list and hard cases go to a person.

  1. 01 · Trigger

    A message arrives

    A customer writes to the support inbox.

  2. 02 · AI step

    It is sorted and matched

    The message is labelled and matched to an approved answer, if one fits.

  3. 03 · Check by a person

    A person decides

    A person approves each draft. Upset customers and unusual cases go straight to a person.

  4. 04 · Action

    The answer is sent

    The approved answer goes out from the support mailbox.

  5. 05 · Record

    The case is recorded

    The message, the answer and who approved it are saved.

A support inbox: messages are sorted and answers are drafted from an approved list. Demonstration with dummy data.

Demonstration with dummy data

Support message card (made-up example)

Message
Sam Example asks how to change a delivery address.
Matched answer
"Changing a delivery address" from the approved answer list.
Needs your decision
The order has already shipped. A person decides what to offer.

Made-up names and numbers. This is not a client result.

Used on: Customer support automation

Example 5

A weekly report is written from website, search and enquiry numbers

A weekly report is written from website, search and enquiry numbers, with the source of every figure named.

  1. 01 · Trigger

    The week ends

    A schedule starts the run.

  2. 02 · AI step

    Numbers are collected and described

    Figures are read from each source and written up in plain words.

  3. 03 · Check by a person

    A person reads it first

    A named person checks the report before anyone else sees it.

  4. 04 · Action

    The report is sent

    The approved report goes to the people who read it.

  5. 05 · Record

    The report is filed

    Each report is saved with its sources.

A weekly report is written from website, search and enquiry numbers. Demonstration with dummy data.

Demonstration with dummy data

Weekly report (made-up example)

Website visits
[number from the website analytics]. Source: the analytics report for the week.
Enquiries
[number from the CRM]. Source: the CRM list for the week.
Note
A figure with no source is not shown. It is listed as missing for a person to check.

Made-up names and numbers. This is not a client result.

Used on: Reporting automation

Example 7

An AI agent with tools, limits and approvals

An AI agent uses tools inside written limits: what it may do alone, what it must ask and what it must never do. A log shows what it did.

  1. 01 · Trigger

    A request arrives

    An email asks for a booking next week.

  2. 02 · AI step

    The agent reads and checks

    It reads the request and looks at the calendar tool. It changes nothing yet.

  3. 03 · Check by a person

    A person approves

    The reply is a draft. The agent cannot send it alone.

  4. 04 · Action

    The reply is sent and the slot held

    After approval, the reply goes out and the time is held.

  5. 05 · Record

    The log is updated

    Each tool used and each decision is recorded.

An AI agent with tools, limits and approvals. Demonstration with dummy data.

Demonstration with dummy data

Agent log (extract)

Step 1
Read the email from Sam Example. Request: a booking next week.
Step 2
Looked up the calendar tool. Two free times found. No change made.
Step 3
Drafted a reply. Marked: needs approval before sending.

Made-up names and numbers. This is not a client result.

Used on: AI agent development · Custom AI development

Example 8

A private assistant answers from a small set of company documents

A private assistant answers a question from a few made-up company documents and shows its source.

  1. 01 · Trigger

    A staff member asks

    "What is our process for an opened item that is returned?"

  2. 02 · AI step

    It looks up the documents

    It finds the relevant passage in the made-up returns policy and answers from it only.

  3. 03 · Check by a person

    The source is shown

    The answer names the document and section. If nothing fits, it says so.

  4. 04 · Action

    The staff member acts

    They follow the process or ask a manager.

  5. 05 · Record

    Questions are logged

    Questions with no answer show which documents are missing.

A private assistant answers from a small set of company documents. Demonstration with dummy data.

Demonstration with dummy data

Assistant answer (made-up example)

Question
What is our process for an opened item that is returned?
Answer
Log the return, inspect the item and ask the manager to approve a refund or exchange.
Source
Returns policy (example document), section 3.

Made-up names and numbers. This is not a client result.

Used on: A private AI assistant on your own documents · Microsoft Copilot consultant

Example 6

One workflow, built four times, on four platforms

The same enquiry workflow as example 1, described on each platform. Platforms differ in how they look and what they cost to run, so the right one depends on your tools.

Demonstration with dummy data

Zapier

A Zap starts from the form entry, runs the sorting and drafting steps, waits for approval and then sends and records.

Zapier consultant page

Make

A scenario starts from the form entry and passes the details through the same steps, with a person approving before the reply.

Make consultant page

n8n

A workflow starts from the form entry and follows the same steps, in the account or environment the customer controls.

n8n consultant page

Microsoft Power Automate

A flow starts from the form entry and follows the same steps, with the approval sent to the person in Microsoft tools.

Microsoft Power Automate consultant page

Read the guide: n8n, Make, Zapier or Power Automate: which one fits your business.